Hi Harry,

I can't comment on the issue with the late notifications, but the jump in
host state from 1/5 SOFT to 1/5 HARD may be due to the fact that Icinga 2
ignores the max_check_attempts option for newly configured (as in not yet
checked) hosts and services. The first check result will immediately be
elevated to a "hard" state.

Regards
Florian

2016-02-23 15:26 GMT+01:00 李寒 <[email protected]>:

> Hi guys,
>
> I run a icinga2 cluster with 4 nodes(2 master, 2 checker), and the
> scheduling behavior is quite strange!
> See my config below. The test-fail service state jumped from 1/5 SOFT, to
> 1/5 HARD, which should be 1/5 SOFT -> 2/5 SOFT -> ... 5/5 SOFT -> 5/5 HARD
> .
> And the notification of test-fail-10 is late. The HARD alert is at
> 1456232652, but the notification is at 1456234216, which is the same time
> with the second time of the test-fail notification.
>
>  # service.conf
>
>  apply Service "test-fail" {
>    max_check_attempts = 5
>    check_interval = 1m
>    retry_interval = 30s
>
>    check_command = "always-fail"
>
>    assign where host.name == "carl2"
>  }
>
>  apply Service "test-fail-10" {
>    max_check_attempts = 3
>    check_interval = 10m
>    retry_interval = 30s
>
>    check_command = "always-fail"
>
>    assign where host.name == "carl2"
>  }
>
>  # zones.conf
>
>  object Endpoint "sindar33a.intra.douban.com" {
>    host = "sindar33a"
>  }
>  object Endpoint "sindar33b.intra.douban.com" {
>    host = "sindar33b"
>  }
>  object Endpoint "sindar33c.intra.douban.com" {
>    host = "sindar33c"
>  }
>  object Endpoint "sindar33d.intra.douban.com" {
>    host = "sindar33d"
>  }
>
>  object Zone "master" {
>      endpoints = [
>          "sindar33a.intra.douban.com",
>          "sindar33b.intra.douban.com",
>      ]
>  }
>
>  object Zone "checker" {
>      endpoints = [
>          "sindar33c.intra.douban.com",
>          "sindar33d.intra.douban.com",
>      ],
>      parent = "master"
>  }
>
>  admin@sindar33a ~ $  tail -F /var/log/icinga2/compat/icinga.log  | grep 
> 'carl2;test'
>  [1456232407] CURRENT SERVICE STATE: carl2;test-fail;UNKNOWN;SOFT;1;
>  [1456232407] CURRENT SERVICE STATE: carl2;test-fail-10;UNKNOWN;SOFT;1;
>  [1456232413] SERVICE ALERT: carl2;test-fail;WARNING;HARD;1;Traceback (most 
> recent call last):
>  [1456232652] SERVICE ALERT: carl2;test-fail-10;WARNING;HARD;1;Traceback 
> (most recent call last):
>  [1456234216] SERVICE NOTIFICATION: 
> lihan-test;carl2;test-fail;WARNING;mail-service-notification;Traceback (most 
> recent call last):;
>  [1456234216] SERVICE NOTIFICATION: 
> lihan-test;carl2;test-fail-10;WARNING;mail-service-notification;Traceback 
> (most recent call last):;
>
>  admin@sindar33b ~ $ tail -F /var/log/icinga2/compat/icinga.log  | grep 
> 'carl2;test'
>  [1456232410] CURRENT SERVICE STATE: carl2;test-fail;UNKNOWN;SOFT;1;
>  [1456232410] CURRENT SERVICE STATE: carl2;test-fail-10;UNKNOWN;SOFT;1;
>  [1456232413] SERVICE ALERT: carl2;test-fail;WARNING;HARD;1;Traceback (most 
> recent call last):
>  [1456232415] SERVICE NOTIFICATION: 
> admin-test;carl2;test-fail;WARNING;mail-service-notification;Traceback (most 
> recent call last):;
>  [1456232652] SERVICE ALERT: carl2;test-fail-10;WARNING;HARD;1;Traceback 
> (most recent call last):
>
>  admin@sindar33c ~ $  tail -F /var/log/icinga2/compat/icinga.log  | grep 
> 'carl2;test'
>  [1456232409] CURRENT SERVICE STATE: carl2;test-fail;UNKNOWN;SOFT;1;
>  [1456232409] CURRENT SERVICE STATE: carl2;test-fail-10;UNKNOWN;SOFT;1;
>  [1456232413] SERVICE ALERT: carl2;test-fail;WARNING;HARD;1;Traceback (most 
> recent call last):
>  [1456232652] SERVICE ALERT: carl2;test-fail-10;WARNING;HARD;1;Traceback 
> (most recent call last):
>
>  admin@sindar33d ~ $  tail -F /var/log/icinga2/compat/icinga.log  | grep 
> 'carl2;test'
>  [1456232408] CURRENT SERVICE STATE: carl2;test-fail;UNKNOWN;SOFT;1;
>  [1456232408] CURRENT SERVICE STATE: carl2;test-fail-10;UNKNOWN;SOFT;1;
>  [1456232413] SERVICE ALERT: carl2;test-fail;WARNING;HARD;1;Traceback (most 
> recent call last):
>  [1456232652] SERVICE ALERT: carl2;test-fail-10;WARNING;HARD;1;Traceback 
> (most recent call last):
>
> Thanks in advance for your help!
>
> Regards
> ​
> --
> Harry Lee  | SA Dept. | Douban Inc.
>
> _______________________________________________
> icinga-users mailing list
> [email protected]
> https://lists.icinga.org/mailman/listinfo/icinga-users
>
>
_______________________________________________
icinga-users mailing list
[email protected]
https://lists.icinga.org/mailman/listinfo/icinga-users

Reply via email to